Character String Input Confidence Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users are often required to manually correct incorrect inputs in input systems, which can be cumbersome, especially in noisy or vibration-prone environments, and there is a need for a method that allows inputs to be made without manual correction.
Innovation Solution
A method that records multiple user inputs to form a character string, performs a confidence analysis by comparing each input with database characters, assigns confidence measures, and outputs character combinations with overall confidence values above a threshold, allowing for automatic correction without user intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual correction of incorrect inputs is required, then input accuracy can be improved, but user convenience and operation speed deteriorate
Solution Approach 1:
The system automatically corrects input errors by comparing user input with database entries and presenting corrected options, eliminating the need for manual correction. The confidence level calculation and automatic suggestion mechanisms enable the system to self-correct without user intervention, resolving the contradiction between input accuracy and user convenience.
Solution Approach 2:
The system provides immediate feedback by displaying confidence levels and alternative corrections to users. This feedback loop allows users to verify automatic corrections or select from suggested alternatives, maintaining high input accuracy while minimizing manual correction effort through informative feedback about system decisions.
2Reliability
If manual correction of inputs is required, then input reliability can be improved, but execution speed and productivity worsen
Solution Approach 1:
The system performs preliminary actions by pre-calculating confidence levels for multiple potential interpretations of user input before final selection is needed. By preparing multiple candidate corrections with their confidence scores in advance, the system ensures reliable input selection while maintaining fast execution speed when the user needs to confirm or select from options.
Solution Approach 2:
The system changes the parameter of confidence level thresholds dynamically. By adjusting the threshold for automatic acceptance versus requiring user selection, the system can balance between input reliability and execution speed based on the specific context, allowing high-confidence inputs to be accepted immediately while lower-confidence inputs trigger verification protocols.
3Measurement precision
If confidence analysis is performed on multiple database entries, then input accuracy improves, but system complexity increases
Solution Approach 1:
The system segments the database into relevant subsets based on the input context and only performs confidence analysis on entries within these subsets. By dividing the large database into manageable segments relevant to the current input, the system maintains high input accuracy through thorough analysis while reducing overall system complexity by avoiding unnecessary comparisons across the entire database.
4Ease of operation
If automatic correction is enabled without user intervention, then operation convenience improves, but measurement precision may worsen
Solution Approach 1:
The system applies partial automatic correction by automatically correcting only those inputs that exceed a high confidence threshold, while requiring user verification for lower-confidence cases. This partial automation approach maintains operation convenience for clear, unambiguous inputs while preserving input accuracy by involving users when the system is less certain about the correct interpretation.
Data Source
Figure 1
Figure 2
Figure 3a~3b
AI summary
The invention relates to a method for inputting and identifying a character string, in which several successively inputted user inputs for forming the character string are identified and a confidence analysis is carried out, in which each user input is compared to a number of characters (7) and each of the compared characters (7) is allocated a value of a confidence measure (22.1, 22.2, 22.3, 22.4) which evaluates the correlation of the user input to said characters. Character combinations (24.1, 24.2, 24.3, 24.4) are formed from the characters (21.1, 21.2, 21.3, 21.4) associated with the user inputs. Global confidence measures (25.1, 25.2, 25.3, 25.4) are determined for the character combinations (24.1, 24.2, 24.3, 24.4) from the confidence measures (22.1, 22.2, 22.3, 22.4) associated with the characters (21.1, 21.2, 21.3, 21.4). Finally, a partial amount of the character combinations (24.1, 24.2, 24.3, 24.4) is emitted in accordance with the global confidence measures (25.1, 25.2, 25.3, 25.4). The invention further relates to a correspondingly designed device (1) for carrying out said method and to a vehicle (9) comprising said type of device (1).